Preventing Health Care-Associated Infection: Development of a Clinical Prediction Rule for Clostridium difficile Infection.

Clostridium Difficile;Nursing;Health Sciences;Nursing

Krapohl, Greta LynnTitler, Marita G. ;

University of Michigan

关键词: Clostridium Difficile;    Nursing;    Health Sciences;    Nursing;   

Others  :  https://deepblue.lib.umich.edu/bitstream/handle/2027.42/89642/krapohlg_1.pdf?sequence=1&isAllowed=y

瑞士|英语

来源: The Illinois Digital Environment for Access to Learning and Scholarship

【 摘 要 】

Introduction: The incidence of Clostridium difficile infection has been steadily rising, growing in virulence, and demonstrating an increase in the severity and morbidity of the disease.A clinical prediction rule (risk score), applied early in, or prior to, hospitalization is a strategy to identify vulnerable patients, target preventative interventions, improve outcomes for Clostridium difficile infection, and translate evidence into clinical practice. Objectives: The purpose of this research was to develop and validate a clinical prediction rule for the risk of Clostridium difficile infection.Methods:Between August 2007 and June 2009, preoperative variables and positive Clostridium difficile assays were collected for adult patients admitted for surgical colectomy from 24 hospitals in Michigan. After performing univariate analysis of 36 preoperative patient risk factors, significant variables associated with Clostridium difficile infection at a p value ≤ .15 were advanced into a binary logistic regression model.The regression coefficients of this model were translated into a weighted scoring system to develop the clinical prediction rule.The receiver operating characteristic curve analysis evaluated the predictive accuracy of the score.Results:2274 patients underwent colectomy and fulfilled inclusion criteria.A total of 55 patients (2.4% overall) developed Clostridium difficile infection.Mechanical ventilation (p=.012) and a history of a transient ischemic attack (p=.042) were independently associated with Clostridium difficile infection.A clinical prediction rule, including the variables from the final model, demonstrated a larger score with an increased patient risk (p ≤ .01).The area under the receiver operating characteristic curve was 0.628 (95% CI .550 -.706).Conclusions:Pulmonary and neurological morbidities emerged as significant preoperative predictive variables of Clostridium difficile infection in this cohort. In contrast to previous studies, bowel preparation, with and without antibiotics, was not associated with an increased risk of CDI.Findings from this study suggest pathogen-directed interventions, such as a clinical prediction rule to quantify the risk factors of Clostridium difficile infection, may offer a promising adjunctive strategy to reduce infection and protect vulnerable patient populations.